Skeleton-Based Posture Classification to Promote Safer Walker-Assisted Gait in Older Adults

2026-04-27

AI Summary

This study addresses the risk of falls among older adults using walkers due to poor posture by proposing an intelligent posture recognition and intervention method based on skeletal keypoints. Through a systematic comparison of geometric feature-based approaches, XGBoost, support vector machines (SVM), and lightweight deep learning models—including a 4-layer CNN and an Encoder-Decoder CNN—in multi-class walker posture classification, the work demonstrates for the first time the feasibility of lightweight machine learning models for high-accuracy real-time applications. Experimental results show that XGBoost achieves over 99% accuracy in binary classification tasks, while the geometric method attains 89.9% accuracy in recognizing eight distinct postures, offering effective technical support for safe human–walker interaction in intelligent assistive devices.

Abstract

Falls among older adults are a significant public health concern, leading to severe injuries, loss of independence, and increased healthcare costs. This study evaluates the effectiveness of various models, including a Geometric approach, XGBoost, SVM, and several deep learning architectures, in classifying walker usage, standing vs. sitting, and posture for smart walkers used. Geometric and XGBoost were the top performers. XGBoost achieved near-perfect training accuracy in binary classification tasks, with 99.84% for walker choice and 99.69% for standing vs. sitting. For posture classification, Geometric approach attained 89.9% accuracy for 8 postures, and XGBoost obtained 99.24% during training for 17 postures. Deep learning models such as the 4-layer CNN and Encoder-Decoder CNN also demonstrated strong performance in binary classification, with accuracies above 98%. This study underscores the potential of machine learning to enhance human-robot interaction in smart walkers, particularly for fall prevention.

Problem

• fall prevention
• skeleton-based posture classification
• smart walkers
• older adults
• gait safety

Innovation

• skeleton-based classification
• XGBoost
• geometric approach
• smart walker
• posture recognition